Multi-population Differential Evolution for RSS based Cooperative Localization in Wireless Sensor Networks with Limited Communication Range

Abstract: This paper presents a novel approach to deal with the cooperative localization problem in wireless sensor networks based on received signal strength measurements. In cooperative scenarios, the cost function of the localization problem becomes increasingly nonlinear and nonconvex due to the heightened interaction between sensor nodes, making the estimation of the positions of the target nodes more challenging. Although most of existing cooperative localization algorithms assure acceptable localization accuracy, their computational complexity increases dramatically, which may restrict their applicability. To reduce the computational complexity and provide competitive localization accuracy at the same time, we propose a localization algorithm based on the differential evolution with multiple populations, opposite-based learning, redirection, and anchoring. In this work, the cooperative localization cost function is split into several simpler cost functions, each of which accounts only for one individual target node. Then, each cost function is solved by a dedicated population of the proposed algorithm. In addition, an enhanced version of the proposed algorithm which incorporates the population midpoint scheme for further improvement in the localization accuracy is devised. Simulation results demonstrate that the proposed algorithms provide comparative localization accuracy with much lower computational complexity compared with the state-of-the-art algorithms.